a 4 THE BO TANK 'AL MAGAZINE. [V oi. xxi. No. 2-11. 



on a high elevation with other Conifers, such as Picea, Abies 

 and Chamcecvparis. The climate there, it is said, is pretty 

 cool all the year through. In a deep valley, each standing- 

 solitary here and there, these trees attain a considerable height 

 of more than fifty metres, and a diameter of about two 

 metres; showing their bare trunks from far off, stretching 

 their clustered foliage towards the apex of the branches. The 

 stem is quite branchless from the base up to the middle to the 

 height of about twenty metres. Together with its branches, 

 it describes an outline of a conical or rather a cylindrical 

 form. Its habit bears much resemblance to Cryptomeria, but 

 has more clustered branchlets and foliage towards the end of 

 the branches. It grows in rather wet places, as is the case 

 with Crvptomeria. The foliage too is very much like that of 

 Cryptomeria, but a little shorter in Taiwania. 



Thus far the two closely resemble each other in the sterile 

 branches. I had myself, perhaps like everybody on the spot, 

 long made the error of thinking that Cryptomeria was found 

 in Formosa, until I first examined the cone of the plant. On 

 glancing over that singular cone, I was at once surprised to 

 find that the new plant must no longer be regarded as belong- 

 ing to Crvptomeria, nor does it seem to be assignable to any 

 known genus. Studying the plant more carefully, I ascertained 

 that the plant should fall into the family of Taxodiineae . 1} 

 Further on I proceeded to find out to what genus it comes 

 nearest, and, if possible, to establish clearly its relation to 

 other forms. It has but one kind of shoots, and, therefore, 

 this no longer comes to Scidopitys. The seed is reversed, by 

 which this differs from Cryptomeria, Taxodium und Glypto- 

 strobus. As Jar as my knowledge extends, the plant comes 

 nearest to Cunninghamia in the structure of its cones, as seen 

 in the arrangement of the seminiferous scales, in the presence 

 of the minute bract,* 3 in the attachment and position of the 



') Engl, and Prantl:— Nat. Pfl.-fam. II. 1, i». 84. 



minghamia i- described as "Bractese distinctse nulla:" or "Bracteae utilise" 



in su.i;. (j t Zuoo. FL Jap. II. pp. (lets'' and in "G. Gordon, The Pinetum, p. 70," 

 without bracts." Bui I convinced myself that there is often, if not always, a 

 minute bract at the ba c e (f each scale. 



